Backend systems work: messaging, data stores, and production debugging.
Production work · Inavitas (energy grid operations)
Event-driven processing for the Outage Management (OMS) and Low Voltage Management (LVM) platforms used by electricity distribution operators. Kafka and BullMQ pipelines decouple incident handling so detection, tracking, and resolution stay responsive under bursty, high-volume event load.
Personal product · shipped iOS app
A quiet, local-first training log for lifters who track reps in reserve. No accounts, no server, no analytics: sessions, sets, and settings stay on the device, with a rest timer backed by local notifications, JSON backup and restore, CSV import from other trackers, and PDF reports.
Personal case study
NestJS microservices for geofence enter/exit detection with PostGIS, transactional outbox publishing to Kafka, and idempotent event logging.
Production work · Inavitas
Hierarchical authorization and complex domain relationships modeled as a graph in RedisGraph and propagated through Redis Pub/Sub, consolidating permission logic that had been scattered across services into a single queryable model.
Production work · Inavitas
Query-plan tuning for high-volume and recursive queries over electrical network topology, outage, and low-voltage datasets, where deep hierarchy traversal drives the latency of operational screens.
Production work · Inavitas
Map-based representation of the electrical distribution network: geospatial data served through PostGIS and GeoServer (WMS, WFS, SLD) and consumed by the platform's operational map layers.
Production work · Inavitas
Standardized logging, error tracking, and dashboards with Grafana and Prometheus across distributed services, plus hands-on incident response spanning backend services, Kafka consumers, and message queues.